Pedestrian tracking from a mobile robot using a laser range finder

Abstract

This paper describes a pedestrian tracking system for a mobile robot. The system detects human position by finding pairs of legs using a laser range finder. It distinguishes between pedestrians and human like static objects by using a static environment map which indicates places of static objects in the environment in which the robot operates. The map is made from the past 200 frames (about 2.6 seconds) of data from a laser range finder. An experiment is reported which confirms the ability of the system to track pedestrians from on board a mobile robot in a daily use environment, even when the robot is in motion. The results also show that the system can distinguish between pedestrians and human like static objects.